Data augmentation in microscopic images for material data mining

TL;DR

This paper introduces a transfer learning approach that combines real and simulated microscopic images to augment training data, significantly reducing data collection costs and improving model performance in material data mining.

Contribution

A novel transfer learning strategy that fuses real and simulated data for effective data augmentation in microscopic image analysis.

Findings

Synthetic images combined with 35% real data outperform full real data training.

Data augmentation reduces data collection time by approximately 65%.

The approach enhances model performance in image segmentation tasks.

Abstract

Recent progress in material data mining has been driven by high-capacity models trained on large datasets. However, collecting experimental data (real data) has been extremely costly since the amount of human effort and expertise required. Here, we develop a novel transfer learning strategy to address small or insufficient data problem. This strategy realizes the fusion of real and simulated data, and the augmentation of training data in data mining procedure. For a specific task of image segmentation, this strategy can generate synthetic images by fusing physical mechanism of simulated images and "image style" of real images. The result shows that the model trained with the acquired synthetic images and 35% of the real images outperforms the model trained on all real images. As the time required to generate synthetic data is almost negligible, this strategy is able to reduce the time cost of real data preparation by roughly 65%.